RESEARCH Open Access

Recessive mutations in ATP8A2 causesevere hypotonia, cognitive impairment,hyperkinetic movement disorders andprogressive optic atrophyHugh J. McMillan1, Aida Telegrafi2, Amanda Singleton2, Megan T. Cho2, Daniel Lelli3, Francis C. Lynn4,5,Julie Griffin6, Alexander Asamoah6, Tuula Rinne7, Corrie E. Erasmus8, David A. Koolen7, Charlotte A. Haaxma9,Boris Keren10, Diane Doummar11, Cyril Mignot10,12,13, Islay Thompson14, Lea Velsher14,Mohammadreza Dehghani15,16, Mohammad Yahya Vahidi Mehrjardi16,17, Reza Maroofian18, Michel Tchan19,20,Cas Simons21, John Christodoulou22, Elena Martín-Hernández23, Maria J. Guillen Sacoto2, Lindsay B. Henderson2,Heather McLaughlin2, Laurie L. Molday24,25, Robert S. Molday24,25 and Grace Yoon26,27*

Abstract

Background: ATP8A2 mutations have recently been described in several patients with severe, early-onset hypotoniaand cognitive impairment. The aim of our study was to characterize the clinical phenotype of patients with ATP8A2mutations.

Methods: An observational study was conducted at multiple diagnostic centres. Clinical data is presentedfrom 9 unreported and 2 previously reported patients with ATP8A2 mutations. We compare their features with3 additional patients that have been previously reported in the medical literature.

Results: Eleven patients with biallelic ATP8A2 mutations were identified, with a mean age of 9.4 years (range2.5–28 years). All patients with ATP8A2 mutations (100%) demonstrated developmental delay, severe hypotoniaand movement disorders, specifically chorea or choreoathetosis (100%), dystonia (27%) and facial dyskinesia (18%).Optic atrophy was observed in 78% of patients for whom funduscopic examination was performed. Symptom onset inall (100%) was noted before 6 months of age, with 70% having symptoms noted at birth. Feeding difficulties werecommon (91%) although most patients were able to tolerate pureed or thickened feeds, and 3 patients requiredgastrostomy tube insertion. MRI of the brain was normal in 50% of the patients. A smaller proportion was noted tohave mild cortical atrophy (30%), delayed myelination (20%) and/or hypoplastic optic nerves (20%). Functional studieswere performed on differentiated induced pluripotent cells from one child, which confirmed a decrease in ATP8A2expression compared to control cells.

(Continued on next page)

* Correspondence: grace.yoon@utoronto.ca26Division of Clinical and Metabolic Genetics, Department of Pediatrics, TheHospital for Sick Children, University of Toronto, 555 University Avenue,Toronto, ON M5G 1X8, Canada27Division of Neurology, Department of Pediatrics, The Hospital for SickChildren, University of Toronto, Toronto, ON, CanadaFull list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 https://doi.org/10.1186/s13023-018-0825-3

(Continued from previous page)

Conclusions: ATP8A2 gene mutations have emerged as the cause of a novel neurological phenotype characterized byglobal developmental delays, severe hypotonia and hyperkinetic movement disorders, the latter being an importantdistinguishing feature. Optic atrophy is common and may only become apparent in the first few years of life,necessitating repeat ophthalmologic evaluation in older children. Early recognition of the cardinal features of thiscondition will facilitate diagnosis of this complex neurologic disorder.

Keywords: ATP8A2, Phospholipid transfer protein, Optic atrophy, Chorea, Choreoathetosis, Dystonia, Developmentaldisabilities, Whole exome sequencing

BackgroundP4-ATPases are a group of proteins that actively transportphospholipids across cell membranes, a process known as‘flipping’ [1, 2]. The main structural phospholipidsare distributed in a non-random manner across thelipid bilayer [3], which is essential for a range offunctions including vesicle trafficking, cellular signalingand neuronal cell survival [4]. Although 14 P4-ATPases(flippases) have been identified, only two genes (ATP8A2and ATP8B1) have been associated with human dis-ease [1, 5].ATP8A2 is highly expressed in the brain, spinal cord,

retina and testis [6, 7]. Mutations in ATP8A2 were ini-tially identified in a family with a clinical phenotype ofcerebellar ataxia, mental retardation and disequilibrium(CAMRQ syndrome) [8]. More recently, ATP8A2 hasbeen linked to a phenotype of intellectual disability, severehypotonia, chorea and optic atrophy without obviousradiographic evidence of cerebellar atrophy [6, 9, 10].We provide a clinical summary of 9 previously unre-

ported patients with ATP8A2 mutations identified viawhole exome sequencing. Detailed clinical informationis also provided for 2 previously reported patients.9

We compare the clinical features of these 11 individualswith three additional published patients [6, 8, 10]. Ex-pression studies of differentiated induced pluripotentcells from one patient revealed decreased ATP8A2RNA expression and protein levels compared to con-trol cells.Our observations confirm that biallelic ATP8A2 muta-

tions cause a distinct clinical phenotype that is character-ized by global developmental delays, severe hypotonia,optic atrophy and hyperkinetic movement disorders.

MethodsPatientsEleven patients from nine families were recruited to par-ticipate in this study. The Research Ethics Board of theHospital for Sick Children approved this study and in-formed consent was obtained from all families accordingto the Declaration of Helsinki. The family of Patient 1consented to a skin biopsy for functional studies to beperformed.

Molecular studiesThree unrelated patients (Patients 1, 2 and 5) had exomesequencing completed at GeneDx (Gaithersburg, MD).Genomic DNA was extracted from whole blood from af-fected children and their parents. Exome sequencingwas performed on exon targets captured using either theAgilent SureSelect Human All Exon (50 Mb) V4 kit orClinical Research Exome kit (Agilent Technologies,Santa Clara, CA). One microgram of DNA from periph-eral blood was sonicated into 300 bp fragments, whichwere then repaired, ligated to adaptors, and purifiedfor subsequent PCR amplification. Amplified productswere then captured by biotinylated RNA library baits in so-lution following the manufacturer’s instructions. BoundDNA was isolated with streptavidin-coated beads andre-amplified. The final isolated products were sequencedusing either the Illumina HiSeq 2500 or 4000 sequencingsystem with either 2 × 100-bp or 2 × 150-bp paired-endreads (Illumina, San Diego, CA). DNA sequence wasmapped to the published human genome build UCSChg19/GRCh37 reference sequence using BWA-Memv0.7.8 [11]. Targeted coding exons and splice junctions ofknown protein-coding RefSeq genes were assessed foraverage depth of coverage with a minimum depth of 10Xrequired for inclusion in downstream analysis. Local re-alignment around insertion-deletion sites was performedusing the Genome Analysis Toolkit v2.3 [12]. Variant callswere generated simultaneously on all sequenced familymembers using either Samtools 0.1.18 or Samtools 0.1.18along with GATK 2.3 HaplotypeCaller [11, 12]. All codingexons and surrounding intron/exon boundaries wereanalyzed. CNVs were called as previously described [13].Automated filtering removed common sequence changes(defined as > 10% frequency present in the 1000 Genomesdatabase). The targeted coding exons and splice junctionsof the known protein-coding RefSeq genes were assessedfor the average depth of coverage and data quality thresholdvalues. Whole exome sequence data for all sequenced fam-ily members was analyzed using GeneDx’s XomeAnalyzer(a variant annotation, filtering, and viewing interface forWES data), which includes nucleotide and amino acid an-notations, population frequencies (NHLBI Exome VariantServer and 1000 Genomes databases), in silico prediction

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 2 of 10

tools and amino acid conservation scores (Mutation Taster,PhyloP, and CADD). Variants were filtered based on in-heritance patterns, gene lists of interest, phenotype,and population frequencies, as appropriate. Resources in-cluding the Human Gene Mutation Database (HGMD),1000 Genomes database, NHLBI Exome Variant Server,OMIM, PubMed, and ClinVar were used to evaluate genesand sequence changes of interest [14–17]. Identified se-quence changes of interest were confirmed in all familymembers by di-deoxy Sanger sequence analysis using anABI3730 (Life Technologies, Carlsbad, CA) and standardprotocols with a new DNA preparation. CNVs were con-firmed in relevant family members by whole genome orexon-focused oligonucleotide array-based comparativegenomic hybridization (Agilent Technologies, Santa Clara,CA), quantitative polymerase chain reaction, or multiplexligation-dependent amplification. Patient 3 underwentexome sequencing at BGI, Denmark with subsequentannotation and interpretation at the Genome Diagnos-tics laboratory in Nijmegen, the Netherlands. Librarieswere prepared using Agilent SureSelect Human AllExon enrichment kit version 5 (Agilent Technologies)and sequenced on an Illumina HiSeq4000 2x150bp.Reads were aligned to hg19 reference genome usingBWA − 0.7.8-r455, variants were called with GenomeAna-lysisTK 3.3–0-g37228af and annotated using an in-housepipeline. Exome sequencing for Patient 4 was carried out inthe molecular laboratory of the Hôpital Pitié-Salpêtrièrein Paris, France. Libraries were prepared from genomicDNA using Roche SeqCap MedExome kits and se-quenced on an Illumina NextSeq 500 2x150bp highoutput (with 12 plexes). Reads were aligned to hg19 ref-erence genome using BWA-mem, variants were calledwith GenomeAnalysisTK-2014.3-17-g0583018 and an-notated using SNPEff-4. Custom scripts were utilizedfor variant filtration and prioritization. Exome sequen-cing for Patients 6 and 7 was carried out at Novogeneusing the Agilent SureSelect V6 enrichment kit with apaired-end (150 bp) protocol at a mean coverage of50X. Reads were aligned to genome assembly hg19with the Burrows-Wheeler Aligner (BWA) (Version0.7.8-r455). Exome sequencing for Patients 8 and 9 wascarried out at the Institute for Molecular Bioscience inQueensland, Australia. Libraries were prepared from gen-omic DNA using Nextera Rapid Capture Exome kits andsequenced on an Illumina HiSeq 2000 to a minimum aver-age coverage of 95X. Reads were aligned to hg19 referencegenome using BWA-mem, variants were called withGATK HaplotypeCaller v3.7 and annotated using SnpEffv4.3 m. Custom scripts were utilized for variant filtrationand prioritization. Patients 10 and 11 were diagnosed at12 de Octubre Hospital in Madrid, Spain and have beenpreviously reported [9]. Detailed exome sequencingmethods are outlined in Additional file 1: Table S1.

Functional studiesHuman pluripotent stem cell differentiationsHuman induced pluripotent stem cells (iPSCs) were gen-erated from control and patient fibroblasts, differenti-ated into endodermal lineage cells, and cell lysates wereused to quantify RNA expression and concentration ofATP8A2 protein.The human iPSCs were generated in house from BJ fi-

broblasts (ATCC) or patient fibroblasts by infecting withSendai virus as outlined in the manufacturer’s protocol(Cytotune 2.0; LifeTechnologies). Cells were plated onMatrigel (Corning) after 48 h and subsequently transi-tioned to mTeSR-E8 media (StemCell Technologies) fromdays 4–7 post-infection. At 3 weeks post-infection, iPSCclones were picked into and maintained in mTeSR-E8 onMatrigel-coated 96-well plates. Clones were passagedusing ReLeSR (Stem Cell Technologies) every 4–6 daysuntil passage 15. Sendai virus transgene expression wasthen analyzed, and found to be absent, using Taqman(Life Technologies) and pluripotency markers assessedby immunostaining and qPCR.hIPSCs were plated onto geltrex (1:100)-coated 12 well

plates at a density of 0.5 × 106 in 10/10 media [DMEM/F12, KOSR, Glutamax, P/S, 10 ng/mL Activin A (E-bio-sciences) and 10 ng/mL herugulin (Tocris) [18]. Differ-entiations began 48 h post-seeding using a modifiedversion of Rezania et al. [19]. Briefly, cells were rinsedwith 1× DPBS (Mg2+ and Ca2+ free) and then basal cul-ture media (MCDB 131 medium, 1.5 g/l sodium bicar-bonate, 1× Glutamax, 1× P/S) with 10 mM final glucose,0.5% BSA, 100 ng/ml Activin A (E-biosciences), and3 μM of CHIR-99021 (Sigma) was added for 1 day only.For the following two days, cells were treated with thesame media without CHIR-99021 compound to generatedefinitive endoderm (Stage 1). On day four, cells werecultured in basal media with 0.5% BSA, 10 mM glucose,0.25 mM ascorbic acid (Sigma) and 50 ng/ml of KGF(R&D Systems) for 2 days to generate primitive gut tube(Stage 2). To produce posterior foregut (Stage 3), cellswere treated for three days with basal media with10 mM final glucose concentration, 2% BSA, 0.25 mMascorbic acid, 50 ng/ml of KGF, 0.25 μM SANT-1(Tocris Biosciences), 1 μM retinoic acid (Sigma), 100 nMLDN193189 (EMD Millipore), 1:200 ITS-X (Gibco), and200 nM TPB (EMD Millipore).

qPCRCells were lysed in Trizol and standard phenol-chloroformextraction was used to isolate RNA as previously described[20]. Following RNA extraction, Superscript III wasused for reverse transcription followed by Taqman onViiA7 384-well thermocycler. Fluorogenic probes usedincluded FAM and the IowaBlack non-fluorescentquencher (PrimeTime; IDT). Primer/probe sequences

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 3 of 10

are: 1) ATP8A2, Primer-1 TGGTTCCTACTGCCTGTTTG; Primer-2 CCTCTTTCCATTGCTATCCCG; ProbeCTTGGTTTCCAGCTCCTGCACCT. 2) TBP, Primer-1GAGAGTTCTGGGATTGTACCG; Primer-2 ATCCTCATGATTACCGCAGC; Probe TGGGATTATATTCGGCGTTTCGGGC.

Protein analysis and western blottingCells were lysed in Laemmeli buffer (2% SDS, 10% gly-cerol, 5% 2-mercaptoethanol, and 0.068 M Tris HCl,pH 6.8) in the absence of bromophenol blue. The solu-tion was boiled for 10 min. The DNA was sheared bypassing the sample through a 26 gauge needle followedby sonication. The sample was then centrifuged for10 min at 1000 g and the supernatant was retained forprotein concentration determination using the BradfordProtein Assay and analysis by SDS gel electrophoresisand Western blotting.For western blotting, proteins were transferred on to

Immobilon FL membranes (Millipore, Bedford, MA)using a BioRad semi-dry transfer apparatus. The blotswere blocked with 1% milk in PBS for 30 min and subse-quently labeled with ATP8A2 polyclonal antibody [21]and actin polyclonal (ab8227, Abcam) antibody as aloading control. The ATP8A2 and actin antibodies werediluted to a concentration of 0.3 mg/ml and 0.2 mg/ml,respectively in PBS containing 0.05% Tween 20 (PBST).The blots were incubated with primary antibodies for1 h, washed with PBST, and subsequently incubated for40 min. With secondary antibody (goat anti-rabbit Igconjugated with horseradish perioxidase (Sigma) diluted1:20,000 in PBST containing 0.1% milk) and washedwith PBST prior to detection by ECL. The film wasscanned on a LiCor imager and the intensity of theATP8A2 labeled bands was quantified using ImageStudio Lite software.

ResultsClinical featuresThe clinical features of individuals with ATP8A2 muta-tions are described for 11 patients ranging from 2.5 yearsto 28 years (Table 1). Patients were of varied ethnicitieswith 8/11 (73%) demonstrating parental consanguinity.All patients exhibited severe hypotonia at or within6 months after birth which persisted, with older childrenand adults being unable to achieve head control and/orsit independently (Table 2). Expressive and receptive lan-guage skills were impaired in all patients. Most children(9/11, 82%) did not develop any meaningful speech, withonly two communicating with mono or di-syllabicwords, or with the aid of pictograms (Table 2). Hyper-kinetic movement disorders, specifically chorea or chor-eoathetosis were present in all patients (Additional file 2:Video S1; Additional file 3: Video S2). Dystonia and/or

facial dyskinesia were noted in a smaller proportion ofpatients (Additional file 4: Video S3). Feeding difficultieswere a significant feature of our cohort and were re-ported in 10/11 (91%) individuals. All 10 patients re-quired modified feeds such as pureed or thickened feeds,and all required significant assistance with feeding.Three patients (Patients 1, 2 and 9) required gastros-tomy tubes due to their risk of aspiration when drinkingthin liquids. Weight was below the third percentile for70% (7/10) of the patients who had a recent weightavailable. Occipitofrontal head circumference (OFC) waswithin the normal range for the majority of patients, and5/11 (45%) had OFC at or below the second percentilefor age. Optic atrophy was noted in 7/9 (78%) patientswho underwent funduscopic examination. The two chil-dren who did not have optic atrophy did not have testingof visual evoked potentials. In one child (Patient 1) whounderwent repeat funduscopic examinations, optic atro-phy was noted to progress in the first few years of life.Initial ophthalmology examination at 10 months was en-tirely normal including funduscopic examination, andthe patient had normal visual fixation and tracking. Re-peat examination at 19 months revealed mild opticnerve pallor. At 26 months she was no longer able tovisually fix or follow and optic nerve atrophy was seen.Visual evoked potentials at 3 years revealed absentcorti-cal responses. At 4.5 years she had severe optic nerve at-rophy (Fig. 1) with optical coherence tomography (OCT)showing selective thinning of the inner retinal layers. Asmaller proportion of patients exhibited ophthalmoplegia(45%) or ptosis (36%) (Additional file 5: Video S4).Seizures were observed in two siblings from a single fam-ily, with the remaining 9/11 patients being seizure-free.Patient 8 did not require treatment with anti-epilepticsand Patient 9 had seizures which were well-controlledwith valproate and carbamazepine. Neither patient hadEEG studies available for review. One patient was noted tohave abnormal brainstem auditory evoked potentials(BAER) responses (Patient 4) consistent with sensori-neural hearing loss. This patient also had abnormal som-atosensory evoked potentials (SSEP) suggesting amyelinopathy involving the dorsal column of the spinalcord. One other patient (Patient 5) was noted to have ab-normal hearing.Nerve conduction studies and electromyography was

performed on 8/11 patients with no evidence of a sen-sorimotor neuropathy observed (Table 3). Single fiberEMG was normal in one patient who had ptosis. Neuro-imaging was performed in all but one patient. MRI ofthe brain revealed no intracranial abnormalities in50% (5/10) patients (Table 3). Non-specific findingsincluded mild cerebral or cortical atrophy (30%), milddelay in myelination (20%), thin corpus callosum(20%) or hypoplastic optic nerves (20%). Additional

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 4 of 10

detailed clinical information is provided for Patients 1and 3 as Additional file 6.

Genetic testingBiallelic ATP8A2 gene mutations were identified in all af-fected individuals, consistent with autosomal recessive in-heritance (Table 4). Carrier parents were asymptomatic.Identified genetic variants included missense, splicing, in-tragenic deletions, and frameshift mutations, which werepredicted to cause loss of function of ATP8A2, and werenot present in the homozygous state in the Exome Aggre-gation Consortium (ExAC) public database or our internalexome databases of unaffected individuals. All variants oc-curred within highly conserved domains, and were pre-dicted to be damaging or possibly damaging by multiplein silico models (SIFT, Polyphen, MetaSVM, MetaSVM,Mutation Taster, PhyloP, and CADD).

Functional studiesInitial studies were unable to detect any ATP8A2 in thecontrol or fibroblasts from Patient 1, due to low ATP8A2expression in this tissue. Subsequent reprogramming of

the fibroblasts into induced pluripotent cells allowed der-ivation of tissues that expressed higher amounts ofATP8A2 to enable study. We found that early endodermderivatives expressed high levels of ATP8A2 and allowedanalyses of RNA expression and ATP8A2 protein expres-sion. To ensure that the patient’s cells could differentiateinto endoderm with equivalent efficiency as control cells,we assessed expression of two endoderm markers, FOXA2and SOX17. Using expression of these markers, the pa-tient’s cells more readily adopted the endoderm fate com-pared to control cells (Supplementary Figure S1). ATP8A2mRNA expression was significantly lower in the patientcompared to control cells and there was diminishedATP8A2 protein in the patient cells, confirming the dele-terious effect of the mutations (Fig. 2). From these studieswe conclude this patient had loss of function mutations inATP8A2, which likely improved endoderm differentiation.

DiscussionDespite the critical role of P4-ATPase proteins in normalcellular functioning, they have only recently been implicatedin human disease. Whole exome sequencing continues to

Table 1 Clinical characteristics of individuals with ATP8A2 mutations

Patients 1 2 3 4 5 6 7 8 9 10 11 Totala

Gender F F F F M F M M F F F

Symptom onset Birth Birth Birth Birth 6 mos 4 mos Birth 6 mos Birth Birth 1 mos

Current age (years) 5y 2.5y 2.7y 6y 5y 9y 15y 16yb 28y 8.5y 5.5y

Gestational age (wk) 40 39 41 40 39 39 40 40 Term 40 39

Birth weight (kg) 4.14 3.57 3.09 4.55 2.62 2.70 2.90 N/a N/a 3.19 2.87

Clinical features

Hypotonia onset Birth Birth Birth Birth Infancy Infancy Birth Infancy Birth Birth Infancy

Hypotonia persists? Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 100%

Muscle weakness Yes Yes Yes N/a No Yes Yes N/a N/a Yes Yes 80%; 8/10

Optic Atrophy Yes N/a Yes Yes Yes No No Yes N/a Yes Yes 78%; 7/9

Ophthalmoplegia No No No Yes No No No Yes Yes Yes Yes 45%

Ptosis No No Yes Yes No No No No Yes No Yes 36%

Hearing Loss No No No Yes Yes No No No No No No 18%

Seizures No No No No No No No Yes Yes No No 18%

Feeding difficulties Yes, G-tube Yes, G-tube Yes No Yes Yes Yes Yes Yes, G-tube Yes Yes 91%

Sleep disturbance Yes Yes No N/a No Yes No No No No Yes 40%; 4/10

Movement disorder

Chorea orchoreoathetosis

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes 100%

Dystonia Yes No Yes Yes No No No No No No Yes 36%

Facial dyskinesia No No No No No No No No No Yes Yes 18%

Current head size (OFC) 25%ile 25%ile 5%ile < 2%ile 10%ile 2%ile 15%ile < 3%ile < 2%ile 25%ile < 2%ile

Current weight 25%ile 83%ile 60%ile N/A < 3%ile < 2%ile < 2%ile < 3%ile < 2%ile < 3%ile < 3%ile

Current length/height 85%ile 90%ile 40%ile N/A 25%ile < 2%ile < 2%ile < 3%ile < 2%ile 5%ile 20%ileaDenominator = 11 unless otherwise indicated; F female, M male, y years old, mos months old, wk weeks, N/a not available, %ile percentile, OFC occipitofrontalhead circumference, NCS/EMG nerve conduction study and electromyography, BAER brainstem auditory evoked response, bDeceased

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 5 of 10

expand our clinical diagnostic capabilities, and in this in-stance has permitted the delineation of a specific neurologicphenotype associated with ATP8A2mutations.An early report described an infant with severe hypo-

tonia and global developmental delay who was found to

have a de novo t(10;13) balanced translocation with thebreakpoint disrupting the coding sequence of a singlegene, ATP8A2 [6]. Although the authors were not ableto quantify ATP8A2 expression, they hypothesized thatthe phenotype may be attributed to haploinsufficiency of

Fig. 1 a Severe bilateral optic atrophy on direct funduscopic examination. b Optical coherence tomography (OCT) reveals relatively thinning of theinner retinal layers suggestive of optic atrophy of those neuronal elements. Outer retinal layers (OPL, ONL) are less affected. RPE = retinal pigmentedepithelium; ONL = outer nuclear layer; OPL = outer plexiform layer, RNFL = retinal nerve fiber layer; GCL = ganglion cell layer; IPL inner plexiform layer

Table 2 Best developmental achievement of individuals with ATP8A2 mutations

Patient Age (years) Language Gross motor Fine motor Feeding G-tube

1 5 Non-verbal Cannot sit Transfers hand-to-hand Requires pureed orthicker feeds

Yes

2 2.5 Babbles Cannot sit Holds objects, not transferring Requires thickenedfoods

Yes

3 2.7 Non-verbal Cannot roll or support head Attempting to grasp Requires pureed orthicker feeds

No

4 6 Non-verbal Cannot sit Cannot grasp No issues No

5 5 Non-verbal Cannot sit Hand grasp Feeding difficulties No

6 9 Non-verbal Impaired Impaired Feeding difficulties No

7 15 Non-verbal Impaired Impaired Feeding difficulties No

8 16a None Impaired Impaired Feeding difficulties No

9 28 None Impaired Impaired Feeding difficulties Yes

10 8.5 Monosyllabic & disyllabic words Cannot support head Holds objects Requires soft foods No

11 5.5 Uses signs, pictograms Cannot support head Attempting to grasp Requires crushedfoods

No

aDeceased

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 6 of 10

the ATP8A2 gene since a mutation on the other allelewas not identified [6]. It was not documented if this pa-tient underwent funduscopic examination, however vis-ual parameters at 1 year of age were reported to benormal. Repeat visual assessment was not reported toascertain if, as for Patient 1 in our case series, optic atro-phy developed later in childhood.Four members of a consanguineous Turkish family

with cerebellar ataxia, mental retardation and disequilib-rium (CAMRQ) syndrome who had previously beenidentified by homozygosity mapping to have shared re-gions of homozygosity on chromosomes 13, 19 and 20underwent whole exome sequencing [8]. A homozygous

c.1128 C > G; p.I376M mutation in ATP8A2 was identi-fied which was predicted to change the secondary struc-ture of the ATP8A2 protein and postulated to be thecause of the CAMRQ syndrome. Phenotypic findingsthat were unique to these patients included truncalataxia with or without quadrupedal gait, dysarthricspeech as well as MRI evidence of mild cerebral andcerebellar atrophy. The presence or absence of optic at-rophy was not reported.One additional child was identified by whole exome

sequencing to have biallelic mutations in ATP8A2 [10].This child demonstrated severe axial hypotonia that wasnoted in the first 6 months of life. At 11 years of age thechild remains non-ambulatory and is capable of speakingsingle-word or short sentences with marked dysarthria.MRI of the brain was normal. He was reported to havechoreoathetosis, dystonia and optic atrophy documentedon funduscopic examination.ATP8A2 is responsible for maintaining a higher con-

centration of phosphatidylserine at the inner surface ofthe phospholipid bilayer [7, 22]. Mouse models ofATP8A2 deficiency have demonstrated that Atp8a2 mu-tations result in impaired axonal transport, axonal loss,failure to thrive and clinical manifestations of neurode-generative disease which are similar to the patients wedescribe [7]. In addition to axonal loss in the spinal cordand retina, mice harbouring Atp8a2 loss of function mu-tations have been found to have axonal degeneration af-fecting peripheral nerves [7]. However, none of thepatients with ATP8A2 mutations in our cohort havedemonstrated any abnormalities on nerve conductionstudies or electromyography (NCS/EMG) (Table 1). Pa-tient 1 underwent a sedated NCS/EMG including con-centric needle EMG of intrinsic foot muscles with noneurogenic changes identified. Due to her fluctuating

Table 4 Genetic characteristics of individuals with ATP8A2 mutations

Patient Ethnicity Alleles Mutations Predicted effect on protein

1 French Canadian, Algerian Compound heterozygote c.1185 + 5G > Adel exons 28–33

Destroys spice donor site in intron 12Partial gene deletion

2 European Ashkenazi NativeAmerican

Compound heterozygote c.1787delAc.321 + 3_321 + 8 delAATGGT

p.Asn596MetfsX – frameshiftDestroys spice donor site in intron 3

3 Turkish Homozygousa c.1756C > T p.Arg586* - premature stop codon

4 Moroccan Homozygousa c.2104 T > C p.Trp702Arg – missense

5 Sri Lankan Homozygousa c.1286A > T p.Lys429Met – missense

6 Iranian Homozygousa c.1474_1662del (del exons 17–18) p.Pro492_Ala554del

7 Iranian Homozygousa c.1474_1662del (del exons 17–18) p.Pro492_Ala554del

8 Lebanese Homozygous c.3188_3196delCTATGGTCC insGAAGAAG p.Thr1063fs - frameshift

9 Lebanese Homozygous c.3188_3196delCTATGGTCC insGAAGAAG p.Thr1063fs - frameshift

10 Spanish Homozygousa c.1287G > T p.Lys429Asn - missense

11 Spanish, Argentinian Compound heterozygote c.1630G > Cc.1873C > T

p.Ala544Pro – missensep.Arg625Trp - missense

aKnown consanguinity

Table 3 Ancillary testing of individuals with ATP8A2 mutations

Patient Nerve conduction MRI brain (age of study, years)

1 Normal Normal; hypoplastic optic nerves(2 yrs., 3 yrs)

2 N/A Normal (1.5 yrs)

3 Normal Mild delay in myelination for age;subcortical white matter volume loss,thin corpus callosum (8 mos, 1.3 yrs)

4 Normal Normal

5 Normal Normal; hypoplastic optic nerves

6 N/A N/A

7 Normal Normal

8 Normal N/A (CT brain =mild cerebral atrophy)

9 Normal Hyperintense signal (T2FLAIR) in opticradiations (12 yo)

10 Normal Delayed myelination for age, mildcerebral atrophy, thin corpuscallosum (6 yrs)

11 N/A Delayed myelination in temporallobes (1.8 yrs)

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 7 of 10

ptosis, Patient 3 underwent NCS with repetitive nervestimulation as well as EMG and stimulated single fiberEMG with no abnormalities demonstrated. The precisereason for this discrepant peripheral nervous systemphenotype between mouse models of ATP8A2 loss offunction mutations and our patients is unknown, anddeserves further study.Previous work has shown that ATP8A2 is important

for function of ectoderm derivatives; however, here wealso demonstrate that ATP8A2 is also expressed in an-other germ layer early in development. In the endoderm,ATP8A2 appears to limit differentiation as patient de-rived induced pluripotent cells with reduced ATP8A2expression were more readily able to adopt the endo-derm fate. In the future, it will be interesting to test howATP8A2 impacts the early embryonic development ofectodermal germ layer in order to understand when itbecomes important for brain development.

ConclusionThis is the largest report of patients with confirmedATP8A2 mutations to date. Despite the variability in mu-tation type and location, the patients in this series showremarkable phenotypic similarity with all having profoundcognitive impairment, severe and persistent hypotonia andchorea or choreoathetosis. The identification of the latterfeature is particularly important in distinguishing patientswith ATP8A2 mutations from the many other geneticcauses of cognitive impairment. Optic atrophy and, lesscommonly, ophthalmoplegia and ptosis can also be seenwhich may be helpful in establishing this diagnosis.

Feeding difficulties and failure to thrive occur frequently,and require dietary modification and occasional interven-tion such as G-tube placement. We suggest that this dis-ease may be under-identified in current clinical practice,and anticipate that increased awareness and diagnosis willlead to a fuller appreciation of the clinical spectrum ofATP8A2-related disorders.

Additional files

Additional file 1: Table S1. Exome sequencing of individuals withATP8A2 mutations. (XLSX 11 kb)

Additional file 2: Video S1. Patient 1 (age 4 years old) demonstrateschorea of her head, upper and lower extremities characteristic of patientswith ATP8A2 mutations. (MOV 7783 kb)

Additional file 3: Video S2. Patient 7 demonstrating chorea of upperand lower limbs. (3GP 64981 kb)

Additional file 4: Video S3. Patient 4 demonstrating dystonic posturingof her arms. (MOV 1273 kb)

Additional file 5: Video S4. Patient 4 demonstrating chorea as well asptosis and ophthalmoplegia. (MOV 2274 kb)

Additional file 6: Supplementary Data - additional clinical histories.Figure S1. Expression of FOXA2 and SOX17 in differentiated cells fromPatient 1 compared to control cells. (PDF 28 kb)

AbbreviationsBAER: Brainstem auditory evoked response; CAMRQ: Cerebellar ataxia, mentalretardation and disequilibrium syndrome; EMG: Electromyography;iPSC: Induced pluripotent stem cells; MRI: Magnetic resonance imaging;NCS: Nerve conduction studies; OCT: Optical coherence tomography;RNA: Ribonucleic acid

Fig. 2 ATP8A2 expression in gut endoderm cells differentiated from patient and control fibroblasts. a ATP8A2 RNA expression levels were determinedby Taqman qPCR; data were normalized to TBP and are expressed as an average +/− SEM for n = 4. b Representative western blots offoregut endodermal cell lysates, differentiated from control or patient induced pluripotent stem cells, and labeled for ATP8A2 and actin (loadingcontrol). c Quantification of ATP8A2 protein expression from western blots expressed as an average +/-SEM for n = 4. P values ** are ≤0.01

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 8 of 10

AcknowledgementsThe authors would like to thank the families for their participation andpermission to publish the results of this study. We thank Dr. J.S. Ponesse forexcellent clinical care provided and Dr. C. Nava for her technical expertise.

FundingThis study was funded in part by Canadian Institute of Health Research GrantMOP 343144 (FCL); Canadian Institute of Health Research Grant MOP PJT148649 (RSM); National Institute of Health EY002422 (RSM).

Availability of data and materialsThe datasets used and/or analyzed during this study are available from thecorresponding author on reasonable request.

Authors’ contributionsHJM and GY wrote the manuscript. DL performed detailed ophthalmologicalevaluations. FCL, LLM RSM performed the functional studies. AT, AS, MC,MGS, LBH, HM, TR, BK, CS provided genetic analysis of whole exomesequencing data. HJM, GY, JG, AA, DK, CE, CAH, CM, DD, LV, IT, EMH, MRD,MYVM, RM, MT, JC provided clinical data necessary to delineate this clinicalphenotype. All authors read and approved the final manuscript.

Ethics approval and consent to participateThe Research Ethics Board of the Hospital for Sick Children approved this studyand informed consent was obtained from all families included in this study.

Consent for publicationAll parents provided written informed consent for the publication of thevideos used in this report.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in publishedmaps and institutional affiliations.

Author details1Division of Neurology, Department of Pediatrics, Children’s Hospital ofEastern Ontario Research Institute, University of Ottawa, Ottawa, ON, Canada.2GeneDx, Gaithersburg, MD, USA. 3Division of Neurology, Department ofMedicine, The Ottawa Hospital, University of Ottawa, Ottawa, ON, Canada.4Diabetes Research Program, Child and Family Research Institute, Vancouver,BC, Canada. 5Department of Surgery and Department of Cellular andPhysiological Sciences, University of British Columbia, Vancouver, BC, Canada.6Weisskopf Child Evaluation Center, Department of Pediatrics, School ofMedicine, University of Louisville, Louisville, KY, USA. 7Department of HumanGenetics, Donders Institute for Brain, Cognition and Behaviour, RadboudUniversity Medical Center, Nijmegen, The Netherlands. 8Department ofNeurology, Donders Center of Neuroscience, Radboud University MedicalCenter, Nijmegen, The Netherlands. 9Department of Neurology, DondersInstitute for Brain, Cognition and Behaviour, Radboud University MedicalCenter, Nijmegen, The Netherlands. 10Assistance Publique Hôpitaux de Paris,Département de Génétique, Groupe Hospitalier, Pitié-Salpêtrière, Paris,France. 11Service de Neuropédiatrie, Hôpital Armand-Trousseau, Paris, France.12Centre de Référence Déficiences Intellectuelles de Causes Rares, GH PitiéSalpêtrière, Paris, France. 13Groupe de Recherche Clinique UPMC DéficienceIntellectuelle de Causes Rares et Autisme GH Pitié-Salpêtrière, Paris, France.14Genetics Program, North York General Hospital, Toronto, ON, Canada.15Medical Genetics Research Centre, Shahid Sadoughi University of MedicalSciences, Yazd, Iran. 16Reproductive Sciences Institute, Shahid SadoughiUniversity of Medical Sciences, Yazd, Iran. 17Diabetes Research Centre, ShahidSadoughi University of Medical Sciences, Yazd, Iran. 18Human GeneticsResearch Centre, Molecular and Clinical Sciences Institute, St George’sUniversity of London, London, UK. 19Department of Genetic Medicine,Westmead Hospital, Westmead, NSW, Australia. 20Sydney Medical School,University of Sydney, Sydney, NSW, Australia. 21Institute for MolecularBioscience, University of Queensland, St Lucia, QLD, Australia.22Neurodevelopmental Genomics Research Group, Murdoch ChildrensResearch Institute and Department of Paediatrics, Melbourne Medical School,University of Melbourne, Melbourne, VIC, Australia. 23Unidad de

Enfermedades Mitocondriales-Metabólicas Hereditarias, Servicio de PediatríaHospital Universitario 12 de Octubre, Universidad Complutense de Madrid,Madrid, Spain. 24Department of Biochemistry and Molecular Biology,University of British Columbia, Vancouver, BC, Canada. 25Department ofOphthalmology and Visual Sciences, Centre for Macular Research, Universityof British Columbia, Vancouver, BC, Canada. 26Division of Clinical andMetabolic Genetics, Department of Pediatrics, The Hospital for Sick Children,University of Toronto, 555 University Avenue, Toronto, ON M5G 1X8, Canada.27Division of Neurology, Department of Pediatrics, The Hospital for SickChildren, University of Toronto, Toronto, ON, Canada.

Received: 22 November 2017 Accepted: 15 May 2018

References1. Coleman JA, Kwok MC, Molday RS. Localization, purification, and functional

reconstitution of the P4-ATPase Atp8a2, a phosphatidylserine flippase inphotoreceptor disc membranes. J Biol Chemi. 2009;284(47):32670–9.

2. Van der Mark VA, Oude Elferink RPJ, Paulusma CC. P4 ATPases: flippases inhealth and disease. Int J Mol Sci. 2013;14:7897–922.

3. Holthius JC, Levine TP. Lipid traffic: floppy drives and a superhighway.Nat Rev Mol Cell Biol. 2005;6:209–20.

4. Andersen JP, Vestergaard AL, Mikkelsen SA, et al. P4-ATPases asphospholipid flippases – structure, function and enigmas. Frontiers Physiol.2016;7:275.

5. Bull LN, Van Eijk MJ, Pawlikowska L, et al. A gene encoding a P-type ATPasemutated in two forms of hereditary cholestasis. Nat Genet. 1998;18:219–24.

6. Cacciagli P, Haddad MR, Mignon-Ravix C, et al. Disruption of the ATP8A2gene in a patient with a t(10;13) de novo balanced translocation and asevere neurological phenotype. Eur J Hum Genet. 2010;18:1360–3.

7. Zhu X, Libby RT, de Vries WN, et al. Mutations in a P-type ATPase genecause axonal degeneration. PLoS Genet. 2012;8(8)

8. Onat OE, Gulsuner S, Bilguvar K, et al. Missense mutation in the ATPase,amino phospholipid transporter protein ATP8A2 is associated with cerebellaratrophy and quadrupedal locomotion. Eur J Hum Genet. 2013;21:281–5.

9. Martin-Hernandez E, Rodriguez-Garcia ME, Camacho A, et al. New ATP8A2gene mutations associated with a novel syndrome: encephalopathy,intellectual disability, severe hypotonia, chorea and optic atrophy.Neurogenetics. 2016;17:259–63.

10. Quintas S, Moldovan O, Proenca dos Santos T, Levy A. New syndrome associatedwith ATP8A2 gene mutations: encephalopathy, intellectual disability, severehypotonia, chorea and optic atrophy; whole exome sequencing role in thediagnosis of new diseases. Eur J Paediatr Neurol. 2017;21:e45–66.

11. Li H, Durbin R. Fast and accurate short read alignment with burrows-wheelertransform. Bioinformatics. 2009;25(14):1754–60.

12. DePristoMA BE, Poplin R, Garimella KV, Maguire JR, Hartl C, Philippakis AA,del Angel G, Rivas MA, Hanna M, McKenna A, Fennell TJ, Kernytsky AM,Sivachenko AY, Cibulskis K, Gabriel SB, Altshuler D, Daly MJ. A framework forvariation discovery and genotyping using next-generation DNA sequencingdata. Nat Genet. 2011;43(5):491–8.

13. Retterer K, Scuffins J, Schmidt D, Lewis R, Pineda-Alvarez D, Stafford A,Schmidt L, Warren S, Gibellini F, Kondakova A, Blair A, Bale S, Matyakhina L,Meck J, Aradhya S, Haverfield E. Assessing copy number from exomesequencing and exome array CGH based on CNV spectrum in a largeclinical cohort. Genet Med. 2015;17(8):623–9.

14. Genomes Project C, Abecasis GR, Auton A, Brooks LD, DePristo MA, DurbinRM, Handsaker RE, Kang HM, Marth GT, McVean GA. An integrated map ofgenetic variation from 1092 human genomes. Nature. 2012;491(7422):56–65.https://doi.org/10.1038/nature11632.

15. Serve EV, NHLBI GO. NHLBI GO exome sequencing project (ESP). Seattle;2015. http://evs.gs.washington.edu/EVS/.

16. OMIM Online Mendelian Inheritance in Man. In: McKusick-Nathans Instituteof Genetic Medicine JHM, National Human Genome Research Institute, editor.An online catalog of human genes and genetic disorders. Baltimore, MD.

17. Landrum MJ, Lee JM, Riley GR, Jang W, Rubinstein WS, Church DM, MaglottDR. ClinVar: public archive of relationships among sequence variation andhuman phenotype. Nucleic Acids Res. 2014;42(Database issue):D980–5.

18. Schultz TC, Young HY, Agulnick AD, et al. A scalable system for productionof functional pancreatic progenitors from human embryonic stem cells.PLoS One. 2012;7(5):e37004.

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 9 of 10

19. Rezania A, Bruin JE, Arora P, et al. Reversal of diabetes with insulin-producing cells derived in vitro from human pluripotent stem cells.Nat Biotechnol. 2014;32(11):1121–33.

20. Krentz NA, Nian C, Lynn FC. TALEN/CRISPER-mediated eGFP knock-in add-on at the OCT4 locus does not impact differentiation of human embryonicstem cells towards endoderm. PLoS One. 2014;9(12):e114275.

21. Coleman JA, Zhu X, Djajadi HR, et al. Phospholipid flippase ATP8A2 isrequired for normal visual and auditory function and photoreceptor andspiral ganglion cell survival. J Cell Sci. 2014;127:1138–49.

22. Vestergaard AL, Coleman JA, Lemmin T, et al. Critical roles of isoleucine-364and adjacent residues in a hydrophobic gate control of phospholipidtransport by the mammalian P4-ATPase ATP8A2. Proc Natl Acad Sci U S A.2014;111(14):1334–43.

McMillan et al. Orphanet Journal of Rare Diseases (2018) 13:86 Page 10 of 10